OFFICES IN Washington D.C. & New Delhi

The Center For Disease Dynamics, Economics & Policy

Vaccinating More People Now by Delaying the Second Shot of the COVID-19 Vaccine Would Save More Lives

CDDEP’s interactive tool shows how different vaccine scenarios affect the transmission of SARS-CoV-2. Modeling suggests vaccinating more individuals now with first shots by delaying second shots could save 20–30% more lives than administering two doses in the winter.

 On Friday, December 11, the United States Food and Drug Administration (FDA) approved the Pfizer vaccine against COVID-19 for emergency use. While the development of the vaccine is a technical achievement to be celebrated, vaccines, per se, do not prevent disease; vaccination does. To actually return to ‘normalcy’, we need to vaccinate a large proportion of the population. The current estimate is that around 60–70 percent of people need to be vaccinated to provide adequate herd immunity. The challenge of distribution starts immediately, and this is further complicated by the need to administer two shots separated by 21–28 days.

Data from the clinical trial of Tozinameran (the name of the Pfizer vaccine) indicated that the vaccine had an efficacy of 52.4% after a single dose and 94.8% after the second dose. The fact that a single dose may be able to provide effectiveness against the disease presents a significant ethical question in the strategy for distribution. The dilemma is caused by the seasonal aspect of transmission. Every year as we enter the winter season, the number of respiratory infections increases, hence the name “Flu Season”. This year, however, it is not the flu that everyone wants to avoid, it’s the novel coronavirus. Several studies, including one by CDDEP, have suggested that SARS-CoV-2 would increase in transmission as temperature and humidity decrease. This is likely due to both biological factors that increase the probability of transmission in the winter and behavioral factors as people are more likely to gather indoors when it is colder and gets darker earlier. Importantly, the inverse of this is also true: as it gets warmer in the spring, the transmission of respiratory viruses slows down. While cases still spread around the US this past summer, the case numbers, hospitalizations, and deaths were nowhere as high as they are now. Thus, the next few months are the crucial period for containing the spread of the virus and avoiding excess morbidity and mortality.

To date, the US federal government has purchased 100 million doses of Tozinameran from Pfizer, and another 200 million doses of mRNA-1273, the vaccine candidate from Moderna that uses the same technology as the Pfizer vaccine and is likely to get approved before the end of the year. This is a total of 300 million doses that theoretically could be available for distribution before May, but because of the 2-shot regimen, we only have enough to vaccinate 150 million people in the short term. Alternatively, second shots could be delayed until the summer, allowing more individuals to be vaccinated sooner. Modeling work we have done suggests that faster distribution of a 50% efficacious vaccine could be as much as 15% more effective in reducing overall levels of infection than a later distribution of a 90% efficacious vaccine. If the vaccine is also able to block transmission and not just prevent disease, and there is reason to think it will at least reduce transmission if not stop it completely, the number of lives saved could be 20%–30% higher. Our interactive tool can help visualize the effects of different vaccine scenarios on the transmission of infectious disease to a generic population.


The caveat of these simulations is that the closer we get to the peak of the winter surge, the less a vaccine can flatten the curve. In other words, the earlier the vaccine is distributed in the epidemic, the greater impact it has on flattening the curve. Across the US, the number of positive cases has increased from around 40,000 a day in August and September to almost 100,000 by the end of October and more than 200,000 a day in the first weeks of December. Yet, despite record numbers of cases, the next couple of months are likely to see even higher case numbers unless we dramatically reduce the spread. While social distancing and face masks can slow transmission, they are not enough. A COVID-19 vaccine distributed in the near term, even if only 50% effective, would be far more effective than a slower distribution of a 90% effective vaccine. Additionally, evidence suggests that the efficacy of the first dose may actually be higher than 50% as data through day seven after the second dose had an efficacy of 90.5%, which increases the potential benefit of vaccinating more individuals early and delaying the administration of a second shot.

Overall, delaying the administration of second shots could be beneficial for four reasons. First, vaccinating more individuals faster during the winter when the transmission is higher would lower overall case numbers and more than offset the lower effectiveness of a single shot. Second, wider distribution and faster falling cases would allow a faster return to normalcy. Third, the durability of the vaccine is unknown, but delaying the booster shot until the summer would likely increase its durability into the fall and potentially next winter, which would help avoid a surge next winter. In addition, the most likely reason for vaccine failure in the short term is antigenic drift. In this process, like influenza, viral mutations allow the virus to evade the immune system and reduce the efficacy of the vaccine. The mRNA technology of these vaccines may allow them to be updated rapidly to combat this problem. Fourth, the reality is that a percentage of individuals will not return or will refuse the second dose. While officials will strive for high compliance, a percentage of individuals will skip the second shot due to hesitancy and misinformation. There is reason to anticipate this as compliance with mandated childhood vaccines is only around 90%, and yearly influenza vaccination rates are only about 50%. Depending on the distribution plans, reserving large amounts for second doses could lead to wasted vaccines. While the clinical trials of the vaccine did not anticipate this — and thus there remains uncertainty as to the effectiveness of only a single dose in the short term, we believe the most ethical choice is to vaccinate as many people as possible and as fast as possible with a single dose before administering the second dose.